@inproceedings{954,
  author = {Ibrahim Akgun and Santiago Vargas and Michael Arkhangelskiy and Andrew Burford and Michael McNeill and Aruna Balasubramanian and Anshul Gandhi and Erez Zadok},
  title = {Predicting Network Buffer Capacity for BBR Fairness},
  abstract = {BBR is a newer TCP congestion control algorithm with promising features, but it can often be unfair to existing loss-based congestion-control algorithms. This is because BBR's sending rate is dictated by static parameters that do not adapt well to dynamic and diverse network conditions. In this work, we introduce BBR-ML, an in-kernel ML-based tuning system for BBR, designed to improve fairness when in competition with loss-based congestion control. To build BBR-ML, we discretized the network condition search space and trained a model on 2,500 different network conditions. We then modified BBR to run an in-kernel model to predict network buffer sizes, and then use this prediction for optimal parameter settings. Our preliminary evaluation results show that BBR-ML can improve fairness when in competition with Cubic by up to 30% in some cases.},
  year = {2022},
  journal = {36th Conference on Neural Information Processing Systems (NeurIPS 2022) Workshop on ML for Systems},
  month = {12},
  url = {https://par.nsf.gov/biblio/10430338},
}